Clinical and Investigative Medicine

 

Factors influencing the estimation of the albumin excretion rate in subjects with diabetes mellitus

Stuart A. Ross, MB, ChB
Gordon H. Fick, PhD
Limor Alima, MD

Clin Invest Med 1997;20(3):152-161

[résumé]

From the Faculty of Medicine, University of Calgary, Calgary, Alta.

(Original manuscript submitted June 4, 1996; received in revised form Mar. 26, 1997; accepted Apr. 1, 1997)

Reprint requests to: Dr. Stuart A. Ross, 238­4411 16 Ave. NW, Calgary AB T3B 0M3

Contents
See also p.140

Abstract

Objective: To evaluate alternative methods of calculating the albumin excretion rate (AER) in the absence of complete and accurate patient documentation, since microalbuminuria in patients with diabetes mellitus is associated with serious complications and since patients often make errors in recording the volume and timing of urine collection, making AER calculations inaccurate.

Design: Prospective study.

Setting: Recruitment sites, including all native reserves, across southern Alberta.

Participants: Population-based group of 1286 subjects with diabetes mellitus participating in the Southern Alberta Study of Diabetic Retinopathy.

Interventions: Timed AERs were measured in the subjects; urinary albumin concentration was measured by radioimmunoassay.

Outcome measures: A formula for the prediction of AER was based on the clinical data from the subjects. Several factors were considered in developing the formula: insulin-using status, weight, sex and urine and serum creatinine concentrations.

Results: A mathematical model for estimation of the AER was developed; incorporation of insulin use, sex and weight provides a more accurate estimate of AER. According to this model, women typically appear to have a lower AER than men and heavier people appear to have a higher AER than people with lower body weight.

Conclusions: The use of mathematical formulae to calculate the AER provides an accurate estimate of the AER, particularly when data related to the volume and timing of urine collection are missing. These formulae will be valuable in large epidemiologic screening programs.

Résumé

Objectif : Évaluer d'autres façons de calculer le taux d'excrétion de l'albumine sans documentation complète et exacte faite par le patient, puisqu'on établit un lien entre la microalbuminurie et de graves complications chez les patients atteints de diabète sucré, et qu'il arrive souvent que des patients font des erreurs en consignant le volume d'urine et l'heure de la collecte, ce qui fausse les calculs du taux d'excrétion de l'albumine.

Conception : Étude prospective.

Contexte : Lieux de recrutement, y compris toutes les réserves autochtones, dans le sud de l'Alberta.

Participants : Groupe démographique de 1286 sujets atteints de diabète sucré participant à l'étude sur la rétinopathie diabétique du sud de l'Alberta.

Interventions : On a mesuré les taux chronométrés d'excrétion de l'albumine chez les sujets; la concentration de l'albumine dans l'urine a été mesurée par radio-immuno-essai.

Mesures des résultats : Une formule de prédiction du taux d'excrétion de l'albumine a été fondée sur les données cliniques tirées des sujets. On a tenu compte de plusieurs facteurs dans l'élaboration de la formule : utilisation d'insuline, masse corporelle, sexe et concentrations de créatinine urinaire et sérique.

Résultats : On a mis au point un modèle mathématique d'estimation du taux d'excrétion de l'albumine. L'intégration de l'utilisation d'insuline, du sexe et de la masse corporelle permet d'estimer avec plus de précision le taux d'excrétion de l'albumine. Selon ce modèle, le taux d'excrétion de l'albumine semble habituellement moins élevé chez les femmes que chez les hommes et plus élevé chez les personnes plus lourdes que chez celles qui ont une masse corporelle inférieure.

Conclusions : L'utilisation de formules mathématiques pour calculer le taux d'excrétion de l'albumine produit une estimation exacte du taux, surtout lorsque des données sur le volume d'urine et le moment de la collecte manquent. Ces formules seront utiles dans le cadre d'importants programmes de dépistage épidémiologique.

[ Top of document ]

Introduction

Microalbuminuria in patients with diabetes mellitus is considered a predictor of hypertension and diabetic renal disease.1­5 Diabetic patients with microalbuminuria can be considered at greater risk of certain diabetic complications and of death than those without albuminuria.4,6­17 Previously, the screening techniques for renal disease included measuring urine albumin by dipstick, a technique that does not identify microalbuminuria. Thus, there is an increasing need to develop easy-to-use screening techniques for microalbuminuria. Because of the difficulties for subjects in collecting a 24-hour urine sample, 2 other collection methods are frequently employed. The first involves a timed overnight urine collection, and the second an early morning urine sample alone. Both methods are considered more practical than a 24-hour sample in a large screening program for microalbuminuria.

To obtain an albumin excretion rate (AER), the subject must note the time when he or she began the collection, the time when the collection was completed and the total volume of the urine collection. If one of these parameters is missing, only an albumin concentration can be determined directly. Several formulae have been developed to enable an observer to calculate excretion rates when not all parameters are available. These formulae range from a simple calculation of albumin concentration and creatinine ratios, to more sophisticated logarithmic techniques.18­20 It is unclear whether other factors, apart from albumin concentration and creatinine level, may influence the calculation of the AER.

The Southern Alberta Study of Diabetic Retinopathy (SASDR) is a major epidemiologic study involving 3646 subjects with diabetes mellitus in southern Alberta, Canada. It is designed to review the prevalence and incidence of retinopathy, nephropathy, hypertension and hyperlipidemia in the subjects. A unique method of recruitment was established; it involves recruitment sites across southern Alberta, including all 6 native reserves in the region. This method of recruitment permits collection of data from a population-based group of subjects with diabetes mellitus, many of whom have never had contact with diabetes health care professionals. By means of a detailed questionnaire and assessment, information is gathered on each of the subjects. This information includes history of diabetes mellitus, vascular complications, family history, insulin-using status, sex and use of health care services. Some basic demographic characteristics of the participants in the SASDR are shown in Table 1. The need for a predictive formula for the AER was motivated by the fact that the subjects in the SASDR did not always record urinary volume and time. The accuracy of estimation of the AER is an important component of the SASDR, which will examine the impact of a variety of factors on changes in the AER and will investigate complications related to microalbuminuria. Thus, it was important to include all potential variables that could affect the prediction of the AER and to measure the effect of these variables, in order to achieve more accurate estimates of the AER. This article outlines the development of a formula that incorporates other parameters that could influence calculation of the AER.

[ Top of document ]

Methods

As part of the SASDR assessment, a timed overnight urine sample was requested. A urine container was mailed to the subjects with an accompanying detailed note describing how to collect the urine sample and what information was to be noted with the sample. All subjects were telephoned before their visit to review the method of urine collection. The subjects were asked to pass urine before going to bed and note the time. They were further instructed to collect any urine that was passed during the night. The next morning, when they arose, they were to pass urine and again note the time. The research technician verified that the patient had collected an overnight timed urine sample, and urine ketones were measured by dipstick. Urine was then forwarded to the laboratory for measurement of albumin concentration.

Albumin concentration was measured by a double-antibody 125I radioimmunoassay (Diagnostic Products Corporation, Los Angeles). This immunoassay method is capable of reading values of 5 µg/mL or higher. The AER was derived from the concentration of albumin, the total time of collection and the volume of overnight urine.

A formula for the prediction of the AER was based on the clinical data from 1286 subjects whose urine had albumin concentrations greater than 5 µg/mL. Insulin use by these subjects, according to sex, is shown in Table 2. Subjects whose urine had albumin concentrations lower than 5 µg/ml were not included in the estimates, since these values are not accurately determined by the radioimmunoassay method.

We considered several factors that could influence the estimate of the AER by affecting renal function or metabolic control, including the concentration of albumin, height, weight, body mass index (BMI), insulin-using status, sex and urine creatinine concentration. These factors are most frequently collected during population surveys of metabolic and renal abnormalities. Also, all of these factors can be obtained by the clinician directly from office records and local laboratories.

A comparison was made of the various models for predicting the AER, to assess their sensitivity in providing accurate estimates. Several formulae were constructed on the basis of the least squares regression method.21 These data were analysed after transforming the measured values with logarithms, as described by Ellis and associates.20 This type of modelling allowed a direct comparison of the derived formulae with the familiar albumin­creatinine ratio, since the ratio is a difference on the log scale. Thus, the albumin­creatinine ratio is of the form:

rate = constant × albumin concentration/creatinine
or
log (rate) = log (constant) + log (albumin concentration) - log (creatinine)

This relationship is called the simple albumin­ creatinine ratio, the only quantity estimated being log (constant). This constant serves to adjust for the difference in units between the AER and the albumin­creatinine ratio.

Given the metabolic effects of insulin use in terms of weight maintenance and renal function, we constructed separate equations for the subjects using insulin and those not using insulin. All candidate models were assessed using regression diagnostics based on residual and probability plots.

[ Top of document ]

Results

The resulting model is referred to as the "complete fitted model." It provides for different relationships, since dependence on creatinine differs between those who use insulin and those who do not. Table 3 reviews 4 models: the simple albumin­creatinine ratio; a fitted albumin­creatinine ratio, similar to that suggested by Ellis and associates;20 a model that allows for the separate contributions of albumin and creatinine (a "fitted albumin and creatinine" model) and the complete fitted model. The variables "gender," "log (weight)," "insulin" and "interaction" were included after "log (albumin)" and "log (creatinine)" to assess their supplementary impact on the AER. The variables are listed in the order of apparent importance, although such ranking should be considered speculative.

It was determined that the nature of dependence of the AER on creatinine was different for subjects using insulin and those not using insulin (p = 0.0406). Accordingly, the formulae included insulin-using status ("insulin"), "log (creatinine)" and their interaction. No other interaction effects were detected. Additive effects for "gender" (p < 0.0001) and "log (weight)" (p < 0.0001) were detected. Although "log (BMI)" was considered a possible factor, "log (weight)" had more predictive usefulness and did not require knowledge of height of the subject.

On the basis of these factors, the following formula was derived for the estimation of the AER:

log (AER) = 1.0496 + 1.0100 log (albumin concentration) - 0.5795 log (creatinine) - 0.1225 gender + 0.5480 log (weight) + 0.1555 insulin - 0.0956 interaction.

The variables "gender" and "insulin" are indicator variables. "Gender" was coded 0 for men and 1 for women; "insulin" was coded 0 for subjects using insulin and 1 for those not using insulin. "Interaction" was coded 0 for insulin users and was the value of "log (creatinine)" for those who did not use insulin, thereby providing for a separate dependence on "log (creatinine)" for the 2 groups.

The validity of these models can be further assessed through their success in correctly predicting a range of clinically relevant intervals (AER < 15, 15­30, 30­70, 70­200 or > 200 µg per minute) (Table 4).

On reviewing these mathematical models, we found that the most elaborate model -- the complete fitted model -- provided the highest percentage of correct predictions, whereas the simple albumin­creatinine ratio gave the lowest percentage. In addition, the complete fitted model made the most substantial gains in agreement over the range of the AER greater than 15 but less than 200 µg per minute. For example, the simple albumin­creatinine ratio had a 45% predictive agreement when the AER was between 15 and 30 µg albumin, whereas the complete fitted model had a 58% agreement. To further support this analysis, a series of logistic regression models were constructed to assess the statistical significance of the independent variables in the correct classification of the AER into these intervals.21 The same variables were statistically significant in predicting the classification of the AER greater than 15 µg per minute versus less than 15 µg per minute and AER greater than 30 µg per minute versus less than 30 µg per minute.

A plot of the formula-based AER versus the volume/time-based AER is shown in Fig. 1. From this graph, and using regression diagnostic techniques, it was determined that some 50 of the subjects did not have an acceptable fit for this model. We reviewed the specific data on each of these 50 subjects and found that all of them had low urinary output rates (less than 20 mL per hour) over the stated collection time, as recorded by the subjects, despite having normal serum creatinine concentrations. We could not determine whether the subjects had inaccurately recorded the timing of the sample or whether, in fact, they were severely dehydrated at the time of the urine collection. Severe hyperglycemia was not noted during their visit for the study review (the day after the overnight urine collection). Fifty other randomly chosen subjects who did fit the regression equation were then considered, and their urine output and parameters were assessed. None of these subjects had a urine excretion rate of less than 40 mL per hour. In no subjects in this cohort were urinary ketones detected.

[ Top of document ]

Discussion

Diabetic nephropathy is recognized as a major cause of illness and death in patients with diabetes mellitus.1,2,4,5,22,23 Nephropathy is often associated with hypertension, and both may remain clinically silent for many years. The detection of large quantities of protein in the urine (macroalbuminuria) heralds a major deterioration of renal function, frequently ending in renal failure. By the time macroalbuminuria has been detected, much of the renal damage is irreversible.

The identification of smaller quantities of protein in the urine (microalbuminuria) has led to a greater understanding of the natural progression of diabetic renal disease. Microalbuminuria may appear many years before the onset of either macroalbuminuria or hypertension.3,7,22 Risk factors contributing to the development and progression of microalbuminuria include poor glucose control, as manifested by elevated HbA1C, and uncontrolled hypertension. Microalbuminuria may, in fact, predict future hypertension and renal impairment.1­3,24­32 It may also predict an association with or future development of vascular complications such as diabetic retinopathy, neuropathy, lipid abnormalities, coronary artery disease and peripheral vascular disease.6­17,22Thus, the ability to detect and accurately measure microalbuminuria has important clinical implications and is an essential component in incidence studies of the vascular complications of diabetes mellitus.

It is recognized that adequate control of blood glucose levels and hypertension may reduce microalbuminuria, or prevent or delay progression to macroalbuminuria.28­32 This fact increases the need to identify all patients who have microalbuminuria so that specific treatments can be initiated early. Increasingly, albumin measurements are being included in major epidemiologic screening studies for the prevalence and incidence of diabetes complications. These studies are important in understanding why some patients develop complications and others do not and the rate at which complications appear. Many aboriginal communities also experience higher-than-average rates of progression of renal disease, and screening methods for microalbuminuria are therefore urgently required in these populations. Screening programs of larger populations have always been difficult because of the need for precise collection procedures, which are not always followed by study subjects. Data may have to be discarded if a minor variation in the collection technique has occurred; this may decrease the benefit of the screening program. During a screening program, basic parameters such as height, weight, sex and insulin use are routinely collected and can thus be easily incorporated into an adjusting formula. The creation of formulae that permit full assessment of screening programs, even when some data are missing, will enhance the ability to carry out screening programs in large populations or in remote areas.

Calculation of the AER requires accurate documentation of the time of collection, the volume of urine and the measurement of albumin concentration. Some controversy has arisen over the ideal collection time for the measurement of microalbuminuria.7,12,18,19,33­43 Traditionally, urine protein has been measured from a 24-hour urine collection, which several studies have suggested to be ideal in assessing microalbuminuria. It appears, however, that albumin excretion may be affected by exercise and food intake; therefore, different AERs may occur during the night, when the subject may be sleeping, than during the day. These concerns have led to the use of the overnight urine collection, which has the benefit of reducing errors, both in measuring the exact time of collection and in remembering to collect all urine samples. Both of these errors would lead to inaccurate estimation of the AER. Unfortunately, as was evident in the SASDR study, subjects collecting urine overnight still make errors in the timing and the volume of their urine collection. This decreases the value of a screening program or long-term follow-up of subjects with possible microalbuminuria. Ketonuria and urinary tract infections may also affect albumin excretion. Although ketones were measured in all subjects, it is not known whether some of the patients had occult or overt urinary tract infections.

Factors that influence metabolic control or renal function may also have an influence on the AER. Thus, use of insulin, sex, weight and serum and urine creatinine concentrations could have an influence on the calculation of the AER.

Recognition of these problems has given rise to several formulae involving the albumin­creatinine ratio that provide an estimate of the AER. As can be seen from the above data, the estimation of the AER appears to involve more than albumin and creatinine values. Incorporation of insulin use, sex and weight provide a more accurate estimate of the AER in a mathematical model. It does appear from this model that, among subjects with the same albumin and creatinine concentrations, women typically have a lower AER than men, and heavier people have a higher AER than those with a lower weight. In addition, subjects who do not use insulin have a higher AER than insulin users over the range of observed creatinine concentrations. It appears to be more accurate to use the square root of the creatinine concentration than the creatinine concentration in calculating the albumin­creatinine ratio.

These data do not provide a ready answer to why sex, weight or insulin use have an effect on albumin secretion. However, several factors that influence albumin excretion have been identified. Insulin resistance and hyperinsulinemia appear to have an effect on the production of albumin through multiple mechanisms, including modification of the mesangium and endothelium structure, and hyperfiltration.44 It is presumed that some of these factors lead to the observed differences in albumin secretion.

This analysis indicates that the AER could be estimated more accurately from the complete fitted formula. As was evident with the 50 subjects, described above, who had a major discrepancy between the measured AER and the complete fitted formula AER, potential errors in either time or volume of urine collection can cause miscalculation of the AER. The use of this complete fitted formula thus allows corroboration of the AER in a large population study of the prevalence and incidence of microalbuminuria. However, it remains possible that the formulae are inaccurate in cases of very low urinary output.

The development of newer urine dipstick methods38,45­48 for measurement of albumin concentration will allow further use of this complete fitted formula to assess the AER. A urine dipstick estimate of albumin concentration, combined with the urine creatinine concentration, may permit increased surveillance of abnormal AER. Although albumin and creatinine concentrations may vary when calculated from a single early-morning or casual urine sample, compared with a timed overnight urine sample, these formulae are still expected to make the AER measurement more accurate. In addition, the ability of specific antihypertensive drugs to modify albumin excretion increases the need to identify patients with microalbuminuria.22,48­60 The formula may also be of some advantage to physicians with office practices who wish to monitor and follow patients with diabetes mellitus with the use of more accurate AER measurements.

One of the most valuable uses of the fitted formula may be in extensive epidemiologic studies. This is particularly important in remote areas, where understanding of renal changes in specific ethnic communities and early detection and adequate treatment are urgently needed. The use of the fitted formula for estimation of microalbuminuria will allow further elaboration of the natural history of this renal abnormality in patients with diabetes mellitus and permit more intensive study of the relationship of microalbuminuria to other complications of diabetes mellitus.

References
  1. Mogensen CE. Microalbuminuria predicts clinical proteinuria and early mortality in maturity-onset diabetes. N Engl J Med 1984;310:356-60.
  2. Mogensen CE, Christensen CK. Predicting diabetic nephropathy in insulin-dependent patients. N Engl J Med 1984;311:89-93.
  3. Viberti GC, Jarrett RJ, Mahmud U, Hill RD, Argyropoulos A, Keen H. Microalbuminuria as a predictor of clinical nephropathy in insulin-dependent diabetes mellitus. Lancet 1982;1:1430-1.
  4. Beatty OL, Ritchie CM, Bell PM, Hadden DR, Kennedy L, Atkinson AB. Microalbuminuria as identified by a spot morning urine specimen in non-insulin-treated diabetes: an eight-year follow-up study. Diabet Med 1995;12:261-6.
  5. Mathiesen ER, Saurbrey N, Hommel E, Parving H-H. Prevalence of microalbuminuria in children with Type I (insulin-dependent) diabetes mellitus. Diabetologia 1986;29:640-3.
  6. Mogensen CE, Vigstrup J, Ehlers N. Microalbuminuria predicts proliferative diabetic retinopathy. Lancet 1985;1:1512-3.
  7. Parving H-H, Hommel E, Mathiesen E, Skott P, Edsberg B, Bahnsen M, et al. Prevalence of microalbuminuria, arterial hypertension, retinopathy and neuropathy in patients with insulin dependent diabetes. BMJ 1988;296:156-60.
  8. Jones SL, Close CF, Mattock MB, Jarrett RJ, Keen H, Viberti GC. Plasma lipid and coagulation factor concentrations in insulin dependent diabetics with microalbuminuria. BMJ 1989;298:487-90.
  9. Dullaart RPF, Dikkeschei LD, Doorenbos H. Alterations in serum lipids and apolipoproteins in male Type I (insulin-dependent) diabetic patients with microalbuminuria. Diabetologia 1989;32:685-9.
  10. Mattock MB, Morrish NJ, Viberti G, Keen H, Fitzgerald AP, Jackson G. Prospective study of microalbuminuria as predictor of mortality in NIDDM. Diabetes 1992;41:736-41.
  11. Krolewski AS, Barzilay J, Warram JH, Martin BC, Pfiefer M, Rand LI. Risk of early-onset proliferative retinopathy in IDDM is closely related to cardiovascular autonomic neuropathy. Diabetes 1992;41:430-7.
  12. Beilin J, Stanton KG, McCann VJ, Knuiman MW, Divitini ML. Microalbuminuria in Type 2 diabetes: an indepent predictor of cardiovascular mortality. Aust N Z J Med 1996;26:519-25.
  13. Torffvit O, Agardh E, Agardh CD. Albuminuria and associated medical risk factors: a cross-sectional study in 476 Type I (insulin-dependent) diabetic patients. Part 1. J Diabet Complications 1991;5:23-8.
  14. Torffvit O, Agardh E, Agardh CD. Albuminuria and associated medical risk factors: a cross-sectional study in 451 Type II (noninsulin-dependent) diabetic patients. Part 2. J Diabet Complications 1991;5:29-34.
  15. Niskanen L, Uusitupa M, Sarlund H, Siitonen O, Voutilainen E, et al. Microalbuminuria predicts the development of serum lipoprotein abnormalities favouring atherogenesis in newly diagnosed Type 2 (non-insulin-dependent) diabetic patients. Diabetologia 1990;33:237-43.
  16. Bell DSH, Ketchum CH, Robinson CA, Wagenknecht LE, Williams BT. Microalbuminuria associated with diabetic neuropathy. Diabetes Care 1992;15:528-31.
  17. Yudkin JS, Forrest RD, Jackson CA. Microalbuminuria as predictor of vascular disease in non-diabetic subjects. Lancet 1988;2:530-33.
  18. Dunn PJ, Jury DR. Random urine albumin: creatinine ratio measurements as a screening test for diabetic microalbuminuria -- a five year follow up. N Z Med J 1990;103:562-4.
  19. Cowell CT, Rogers S, Silink M. First morning urinary albumin concentration is a good predictor Of 24-hour urinary albumin excretion in children with Type I (insulin-dependent) diabetes. Diabetologia 1986;29:97-9.
  20. Ellis D, Coonrod BA, Dorman JS, Kelsey SF, Becker DJ, Avner ED, et al. Choice of urine sample predictive of microalbuminuria in patients with insulin-dependent diabetes mellitus. Am J Kidney Dis 1989;13:321-328.
  21. Matthews DE, Farewell VT. Using and understanding medical statistics. Basel (Switzerland): Karger; 1988.
  22. Bennett PH, Haffner S, Kasiske BL, Keane WF, Mogensen CE, Parving H-H, et al. Screening and management of microalbuminuria in patients with diabetes mellitus: recommendations to the Scientific Advisory Board of the National Kidney Foundation from an Ad Hoc Committee of the Council on Diabetes Mellitus of the National Kidney Foundation. Am J Kidney Dis 1995;25:107-12.
  23. Batlle D. Attenuation of nephrosis and its lipid complications. Can J Diagn 1996;(suppl):14­6.
  24. Vigstrup J, Mogensen CE. Proliferative diabetic retinopathy: at risk patients identified by early detection of microalbuminuria [abstract]. Acta Ophthalmol 1985;63(5):530-4.
  25. Berglund J, Lins PE, Adamson U, Lins L-E. Microalbuminuria in long-term insulin-dependent diabetes mellitus. Acta Med Scand 1987;222:333-8.
  26. Mogensen CE, Schmitz O. The diabetic kidney: from hyperfiltration and microalbuminuria to end-stage renal failure. Med Clin North Am 1988;72:1465-92.
  27. Schmitz A, Vaeth M. Microalbuminuria: a major risk factor in non-insulin-dependent diabetes. A 10-year follow-up study of 503 patients. Diabet Med 1988;5(2):126-34.
  28. D'Antonio JA, Ellis D, Doft BH, Becker DJ, Drash AL, Kuller LH, et al. Diabetes complications and glycemic control. The Pittsburgh prospective insulin-dependent diabetes cohort study status report after 5 yr of IDDM. Diabetes Care 1989;12:694-700.
  29. Viberti GC, Walker JD. Diabetic nephropathy: etiology and prevention. Diabetes Metab Rev 1988;4:147-62.
  30. Kikkawa R, Haneda M, Togawa M, et al. Microalbuminuria associated with a rise in blood pressure in non-insulin-dependent diabetes. J Diabet Complications 1989;3(2):99-102.
  31. Chase HP, Garg SK, Harris S, Hoops SL, Marshall G. High-normal blood pressure and early diabetic nephropathy. Arch Intern Med 1990;150:639-41.
  32. Microalbuminuria Collaborative Study Group. Microalbuminuria in Type I diabetic patients. Prevalence and clinical characteristics. Diabetes Care 1992;15:495-501.
  33. Gatling W, Knight C, Hill RD. Screening for early diabetic nephropathy: Which sample to detect microalbuminuria? Diabet Med 1985;2:451-55.
  34. Eshoj O, Feldt-Rasmussen B, Larsen ML, Mogensen EF. Comparison of overnight, morning and 24-hour urine collections in the assessment of diabetic microalbuminuria. Diabet Med 1987;4:531-3.
  35. Gatling W, Knight C, Mullee MA, Hill RD. Microalbuminuria in diabetes: a population study of the prevalence and an assessment of three screening tests. Diabet Med 1988;5:343-7.
  36. Damsgaard EM, Mogensen CE. Microalbuminuria in elderly hyperglycaemic patients and controls. Diabet Med 1986;3:430-5.
  37. Cohen DL, Close CF, Viberti GC. The variability of overnight urinary albumin excretion in insulin-dependent diabetic and normal subjects. Diabet Med 1987;4:437-40.
  38. Baba T, Murabayashi S, Tomiyama T, et al. Microalbuminuria and posture [letter]. Lancet 1988;2:970.
  39. Sochett E, Daneman D. Screening tests to detect microalbuminuria in children with diabetes. J Pediatr 1988;112(5):744-8.
  40. Stehouwer CD, Fischer HRA, Hackeng WHL, den Ottolander GJH. Identifying patients with incipient diabetic nephropathy. Should 24-hour urine collections be used? Arch Intern Med 1990;150:373-5.
  41. Ciavarella A, Silletti A, Forlani G, Morotti L, Borgnino LC, D'Apote M, et al. A screening test for microalbuminuria in Type 1 (insulin-dependent) diabetes. Diabetes Res Clin Pract 1989;7:307-12.
  42. Howey JEA, Browning MCK, Fraser CG. Biologic variation of urinary albumin: consequences for analysis, specimen collection, interpretation of results, and screening programs. Am J Kidney Dis 1989;13:35-7.
  43. Marshall SM. Screening for microalbuminuria: Which measurement? Diabet Med 1991;8:706-11.
  44. Laplante L. Detecting microalbuminuria in your office. Can J Diagn 1996;(suppl):9-10.
  45. Watts GF, Hodgson B, Morris RW, Shaw KM, Polak A. Side-room tests to screen for microalbuminuria in diabetes mellitus. Diabet Med 1988;5:298-303.
  46. Bangstad HJ, Try K, Dahl-Jorgensen K, Hanssen KF. New semiquantitative dipstick test for microalbuminuria. Diabetes Care 1991;14:1094-7.
  47. Gilbert RE, Akdeniz A, Jerums G. Semi-quantitative determination of microalbuminuria by urinary dipstick. Aust N Z J Med 1992;22:334-7.
  48. de Grauw WJ, de Lisdonk EH, van den Hoogen HJ. Screening for microalbuminuria in type 2 diabetic patients: the evaluation of a dipstick test in general practice. Diabet Med 1995;12:657-63.
  49. Marre M, Chatellier G, Leblanc H, Guyenne TT, Menard J, Passa P. Prevention of diabetic nephropathy with enalapril in normotensive diabetics with microalbuminuria. BMJ 1988;297:1092-5.
  50. Marre M, Leblanc H, Suarez L, Guyenne TT, Menard J, Passa P. Converting enzyme inhibition and kidney function in normotensive diabetic patients with persistent microalbuminuria. BMJ 1987;294:1448-52.
  51. Winocour PH, Waldek S, Anderson DC. Converting enzyme inhibition and kidney function in normotensive diabetic patients with persistent microalbuminuria [letter]. BMJ 1987;295:391.
  52. Mimran A, Insua A, Ribstein J, Monnier L, Bringer J, Mirouze J. Contrasting effects of captopril and nifedipine in normotensive patients with incipient diabetic nephropathy. J Hypertens 1988;6:919-23.
  53. Insua A, Ribstein J, Mimran A. Comparative effect of captopril and nifedipine in normotensive patients with incipient diabetic nephropathy. Postgrad Med J 1988;64:59-62.
  54. Jenkins AJ, Steele JS, Janus ED, Best JD. Increased plasma apolipoprotein(a) levels in IDDM patients with microalbuminuria. Diabetes 1991;40:787-90.
  55. Mathiesen ER, Hommel E, Giese J, Parving H-H. Efficacy of captopril in postponing nephropathy in normotensive insulin dependent diabetic patients with microalbuminuria. BMJ 1991;303:81-6.
  56. Winocour PH, Bhatnagar D, Ishola M, Arrol S, Durrington PN. Lipoprotein (a) and microvascular disease in Type 1 (insulin-dependent) diabetes. Diabet Med 1991;8:922-7.
  57. Dhanesha U, Miles D, Bradford N, Gilchrist J, Weatherill J. Visual function and microalbuminuria [letter]. Diabet Med 1992;9:95.
  58. Romanelli G, Giustina A, Bossoni S, Caldonazzo A, Cimino A, Cravarezza P, et al. Short-term administration of captopril and nifedipine and exercise-induced albuminuria in normotensive diabetic patients with early-stage nephropathy. Diabetes 1990;39:1333-8.
  59. Alzaid AA. Microalbuminuria in patients with NIDDM: an overview. Diabetes Care 1996;19:79-89.
  60. Parving H-H, Tarnow L, Rossing P. Renal protection in diabetes: an emerging role for calcium antagonists. J Hypertens 1996;14(suppl 4):s21-s25.

| CIM: June 1997 / MCE : juin 1997 |
CMA Webspinners / >